Schwarz’s criterion, also known as the Bayesian Information Criterion or BIC, is commonly used for model selection in logistic regression due to its simple intuitive formula. For tests of nested hypotheses in independent and identically distributed data as well as in Normal linear regression, previous results have motivated use of Schwarz’s criterion by its consistent approximation to the Bayes factor (BF), defined as the ratio of posterior to prior model odds. Furthermore, under construction of an intuitive unit-information prior for the parameters of interest to test for inclusion in the nested models, previous results have shown that Schwarz’s criterion approximates the BF to higher order in the neighborhood of the simpler nested model. This paper extends these results to univariate and multivariate logistic regression, providing approximations to the BF for arbitrary prior distributions and definitions of the unit-information prior corresponding to Schwarz’s approximation. Simulations show accuracies of the approximations for small samples sizes as well as comparisons to conclusions from frequentist testing. We present an application in prostate cancer, the motivating setting for our work, which illustrates the approximation for large data sets in a practical example.
Background In Eurotransplant kidney allocation system (ETKAS) candidates can be considered unlimitedly for repeated re‐transplantation. Data on outcome and benefit is indeterminate. Methods Retrospective 15‐year patient and graft outcome data analysis from 1464 recipients of a third or fourth or higher sequential deceased donor renal transplantation (DDRT) from 42 transplant centers. Results Repeated re‐DDRT recipients were younger (mean 43.0 vs. 50.2 years) compared to first DDRT recipients. They received grafts with more favorable HLA‐matches (89.0% vs. 84.5%) but thereby no statistically significant improvement of patient and graft outcome was found as comparatively demonstrated in 1st DDRT. In the multivariate modelling accounting for confounding factors, mortality and graft loss after 3rd and ≥4th DDRT (p<0.001 each) and death with functioning graft (DwFG) after 3rd DDRT (p=0.001) were higher as compared to 1st DDRT. The incidence of primary non‐function (PNF) was also significantly higher in re‐DDRT (12.7%) than in 1st DDRT (7.1%; p<0.001). Conclusions Facing organ shortage, increasing waiting time, and considerable mortality on dialysis we question the current policy of repeated re‐DDRT. The data from this survey proposes better HLA‐matching in first DDRT and second DDRT and careful selection of candidates, especially for ≥4th DDRT.